An exemplary method for online character recognition of East Asian characters includes acquiring time sequential, online ink data for a handwritten East Asian character, conditioning the ink data to produce conditioned ink data where the conditioned ink data includes information as to writing sequence of the handwritten East Asian character and extracting features from the conditioned ink data where the features include a tangent feature, a curvature feature, a local length feature, a connection point feature and an imaginary stroke feature. Such a method may determine neighborhoods for ink data and extract features for each neighborhood. An exemplary Hidden Markov Model based character recognition system may use various exemplary methods for training and character recognition.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for online character recognition of East Asian characters, implemented at least in part by a computing device, the method comprising: acquiring time sequential, online ink data for a handwritten East Asian character; conditioning the ink data to produce conditioned ink data where the conditioned ink data comprises information as to writing sequence of the handwritten East Asian character; extracting features from the conditioned ink data wherein the features comprise a tangent feature, a curvature feature, a local length feature, a connection point feature and an imaginary stroke feature; and training a Hidden Markov Model based character recognition system using the extracted features.
2. The method of claim 1 wherein the conditioning generates ink data frames wherein the ink data frames comprise a uniform length, as defined between two ink data points.
3. The method of claim 1 wherein the conditioning generates a series of contiguous ink data frames wherein the series comprises real stroke frames and imaginary stroke frames.
4. The method of claim 1 further comprising determining neighborhoods that comprise successive ink data frames.
5. The method of claim 4 wherein the determining neighborhoods comprises: determining a turning angle between two adjacent ink data frames; determining at least one other turning angle between the two adjacent ink data frames; determining a cumulative angle based on the turning angle and the at least one other turning angle; and comparing the cumulative angle to a predetermined threshold to decide if the two adjacent ink data frames belong to a same neighborhood.
6. The method of claim 1 comprising determining neighborhoods of ink data and extracting a local length feature for each neighborhood.
7. The method of claim 1 wherein the extracting a tangent feature comprises, for a x, y Cartesian coordinate system, determining a Δx value and a Δy value for each ink data frame.
8. The method of claim 1 wherein the extracting a curvature feature comprises determining a sine value and a cosine value for an angle between two adjacent ink data frames.
9. The method of claim 1 further comprising selecting an East Asian character as associated with the handwritten East Asian character using the Hidden Markov Model based character recognition system.
10. A method for online character recognition of East Asian characters, implemented at least in part by a computing device, the method comprising: acquiring time sequential, online ink data for a handwritten East Asian character; conditioning the ink data to produce conditioned ink data where the conditioned ink data comprises ink data frames and information as to writing sequence of the handwritten East Asian character; determining neighborhoods of ink data frames wherein the determining neighborhoods comprises: determining a turning angle between two adjacent ink data frames; determining at least one other turning angle between the two adjacent ink data frames; determining a cumulative angle based on the turning angle and the at least one other turning angle; and comparing the cumulative angle to a predetermined threshold to decide if the two adjacent ink data frames belong to the same neighborhood; and applying a Hidden Markov Model based character recognition system to recognize the handwritten East Asian character.
11. The method of claim 10 further comprising determining a local length feature for each of the neighborhoods.
12. The method of claim 10 further comprising determining one or more connection point features for each of the neighborhoods wherein the one or more connection point features comprises at least one member selected from a group consisting of cross connection points and “T” connection points.
13. The method of claim 10 wherein the conditioned ink data comprises ink data frames for real strokes and ink data frames for imaginary strokes.
14. A computing device comprising: a processor; a user input mechanism; a display; and control logic implemented at least in part by the processor to recognize an online, handwritten East Asian character based on a character recognition algorithm that uses a Hidden Markov Model (HMM) and to extract features from online, handwritten East Asian character ink data, the extracted features comprising a tangent feature, a curvature feature, a local length feature, a connection point feature and an imaginary stroke feature.
15. The computing device of claim 14 , wherein the control logic is further implemented to uniformly sample the ink data and to generate ink data frames of uniform length.
16. The computing device of claim 14 , wherein the control logic is further implemented to generate a series of contiguous ink data frames from the character ink data, wherein the series of ink data frames comprises real stroke frames and imaginary stroke frames.
17. The computing device of claim 14 , wherein the control logic is further implemented to generate ink data frames from the character ink data and to determine, for a x, y Cartesian coordinate system, a Δx value and a Δy value for each ink data frame.
18. The computing device of claim 14 , wherein the control logic is further implemented to generate ink data frames from the character ink data and to determine a sine value and a cosine value for an angle between two adjacent ink data frames.
19. The computing device of claim 14 , wherein the control logic is further implemented: to generate ink data frames from the character ink data; and to determine neighborhoods, wherein each neighborhood comprises successive ink data frames.
20. The computing device of claim 14 , wherein the computing device comprises a mobile phone.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
June 29, 2007
July 5, 2011
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.